Wim Taymans [Mon, 20 Apr 2009 12:01:01 +0000 (14:01 +0200)]
-launch: connect to deep-notify with right name
Connect to the right signal name with - instead of _.
Jan Schmidt [Fri, 24 Apr 2009 21:06:19 +0000 (22:06 +0100)]
info: Support new printf extensions in glibc 2.10
The printf extension mechanism changed in glibc 2.10, and the older
register_printf_function is deprecated. Detect and use the new
mechanism where available.
Jan Schmidt [Mon, 20 Apr 2009 11:25:57 +0000 (12:25 +0100)]
docs: Fix a typo in the release script
Jan Schmidt [Mon, 11 May 2009 20:11:49 +0000 (21:11 +0100)]
Back to development -> 0.10.23.1
Jan Schmidt [Sun, 10 May 2009 21:41:04 +0000 (22:41 +0100)]
Release 0.10.23
Jan Schmidt [Sun, 10 May 2009 21:38:45 +0000 (22:38 +0100)]
Update .po files
Jan Schmidt [Wed, 6 May 2009 15:10:11 +0000 (16:10 +0100)]
0.10.22.4 pre-release
Wim Taymans [Fri, 24 Apr 2009 17:36:22 +0000 (19:36 +0200)]
GstBin: set PENDING_STATE correctly
Set the pending state correctly when we are going to perform an async
state_continue on the bin.
Fixes #580121
Jan Schmidt [Tue, 21 Apr 2009 19:50:55 +0000 (20:50 +0100)]
0.10.22.3 pre-release
Jan Schmidt [Tue, 21 Apr 2009 21:12:04 +0000 (22:12 +0100)]
Automatic update of common submodule
From b3941ea to 6ab11d1
Tim-Philipp Müller [Fri, 17 Apr 2009 14:46:52 +0000 (15:46 +0100)]
win32: define __MSVCRT_VERSION__ when compiling with MingW, for __stat64
Need to define this when using MingW, so that the includes provide
__stat64 and friends. We need at least Windows XP SP2 for this.
Fixes #568632.
Stefan Kost [Thu, 16 Apr 2009 19:26:00 +0000 (22:26 +0300)]
gstdebug: compete stubs. Fixes #579177.
Avoid defines when including gstinfo.h ourself and complete stubs. Sync stub
returns with the defines.
Yaakov Selkowitz [Fri, 17 Apr 2009 10:44:11 +0000 (11:44 +0100)]
configure.ac: fork() during registry scanning is unsafe on Cygwin
Fixes #555978.
Yaakov Selkowitz [Fri, 17 Apr 2009 10:39:59 +0000 (11:39 +0100)]
gst_init: relocatability is unnecessary on Cygwin
See #555978.
Brian Cameron [Fri, 17 Apr 2009 09:11:21 +0000 (10:11 +0100)]
gstinfo: don't assume G_HAVE_ISO_VARARGS implies ISO C99
Makes headers C++ clean, esp. with the Sun compilers.
Fixes #567692.
Tim-Philipp Müller [Fri, 17 Apr 2009 08:17:40 +0000 (09:17 +0100)]
GstPlugin: fix compilation if both HAVE_WIN32 and HAVE_SIGACTION are defined
Move _gst_plugin_fault_handler_is_setup into the ifdef block where it's
used. Fixes #578201.
Wim Taymans [Thu, 16 Apr 2009 10:01:50 +0000 (12:01 +0200)]
-launch: disable CLOCK_LOST message handling
Disable the handling of the CLOCK_LOST messages until we fixed and released the
elements (rtspsrc) that break when we quickly PAUSE/PLAY the pipeline.
Fixes #579127
Jan Schmidt [Wed, 15 Apr 2009 21:24:45 +0000 (22:24 +0100)]
release: Bump version to 0.10.22.2 for pre-release
Jan Schmidt [Wed, 15 Apr 2009 23:08:20 +0000 (00:08 +0100)]
win32: Update win32 build files
Jan Schmidt [Wed, 15 Apr 2009 22:27:31 +0000 (23:27 +0100)]
po: Update translations from TP
Jan Schmidt [Wed, 15 Apr 2009 21:17:10 +0000 (22:17 +0100)]
ChangeLog: regenerate changelog with the gen-changelog script
Jan Schmidt [Wed, 15 Apr 2009 22:26:13 +0000 (23:26 +0100)]
docs: remove errant gtk-doc comment marker triggering a warning
Jan Schmidt [Wed, 15 Apr 2009 23:02:07 +0000 (00:02 +0100)]
paramspecs: revert gst_param_spec_is_mutable() for release
Revert the gst_param_spec_is_mutable API for this release so we can
discuss it a bit further first.
Stefan Kost [Wed, 15 Apr 2009 20:33:20 +0000 (23:33 +0300)]
logging: fix unused variable warning when disabling debug logs.
The var was NULL anyway, bacause of the ifdefs there, the message makes no
sense including it.
Stefan Kost [Wed, 15 Apr 2009 20:12:11 +0000 (23:12 +0300)]
gstdebug: show enabled/disabled in configure and fix build for disabled
When its disabled, we poison some symbols to force a build error if they are
used. Dunno how useful this acually is, but we need to disable the poisoning
when we include this ourself. Also don't define some of the dummies, as they
are getting replaced with defines and that creates code that does not compile.
Sebastian Dröge [Wed, 15 Apr 2009 17:58:34 +0000 (19:58 +0200)]
Use g_once_init_*() instead of GOnce for the enum types
Sebastian Dröge [Wed, 15 Apr 2009 11:05:16 +0000 (13:05 +0200)]
staticpadtemplate: Update docs of gst_static_pad_template_get_caps ()
gst_static_pad_template_get_caps () actually returns a reference to the
caps and it's cleaner to unref them after usage. The core will, however,
always hold a reference to the caps so this didn't result in a memory
leak.
Stefan Kost [Tue, 14 Apr 2009 19:32:21 +0000 (22:32 +0300)]
docs: use real <note> tags as they look nice in new gtk-doc
David Schleef [Tue, 14 Apr 2009 19:20:37 +0000 (12:20 -0700)]
Fix locking in gst_param_spec_is_mutable
Stefan Kost [Tue, 14 Apr 2009 19:07:38 +0000 (22:07 +0300)]
controller: factor out duplicated code and add a description for it.
Also fix typo in the tests while reviewing them.
Wim Taymans [Tue, 14 Apr 2009 17:12:52 +0000 (19:12 +0200)]
docs: add simple doc blurb
Wim Taymans [Tue, 14 Apr 2009 17:11:44 +0000 (19:11 +0200)]
paramspecs: add note about racyness
Add a note about potential racyness in _is_mutable().
LRN [Tue, 14 Apr 2009 08:32:07 +0000 (10:32 +0200)]
info: use mutex to do console colors on windows
Use a static mutex to keep the console colors and context together when
debugging with colors on Windows.
Fixes #517231.
Tim-Philipp Müller [Mon, 13 Apr 2009 13:27:49 +0000 (14:27 +0100)]
docs: add Since: tags to gtk-doc chunks for new param spec API
And, for our release manager, the in-commit-message keywords
for the previous commit:
API: GST_PARAM_MUTABLE_READY
API: GST_PARAM_MUTABLE_PAUSED
API: GST_PARAM_MUTABLE_PLAYING
API: gst_param_spec_is_mutable
David Schleef [Fri, 20 Feb 2009 19:09:19 +0000 (11:09 -0800)]
Add param spec flags for when a property can be changed
Adds GST_PARAM_MUTABLE* flags to indicate in which states a
property can be changed and take effect. Fixes #571559
Wim Taymans [Fri, 10 Apr 2009 12:15:36 +0000 (14:15 +0200)]
-launch: handle clock-lost messages
When we receive a clock-lost message, we need to select a new clock in the
pipeline by setting the pipeline to PAUSED and back to PLAYING.
Olivier Crete [Thu, 9 Apr 2009 16:27:21 +0000 (18:27 +0200)]
tee: add property to control the alloc pad
Add a property to control the pad used for proxying the buffer_alloc function on
the sinkpad.
Fixes #577891.
Wim Taymans [Thu, 9 Apr 2009 09:51:43 +0000 (11:51 +0200)]
bin: always mark pending_async_done
When we get an ASYNC_DONE message when a state change was busy, set the
pending_async_done flag so that after the state change completes, the bin can
check if all async elements are finished. Don't only do this for the bin itself
but for all elements.
This fixes some bins in bins that simulate async state changes by posting ASYNC
messages (such as sdpparse in uridecodebin/playbin2).
Wim Taymans [Thu, 9 Apr 2009 09:42:48 +0000 (11:42 +0200)]
info: fix compilation, %08x needs an unsigned int
%08x needs an unsigned int, so give it that.
Tim-Philipp Müller [Mon, 6 Apr 2009 00:27:26 +0000 (01:27 +0100)]
API: add FIXME and DUMPMEM log levels and convenience macros
Two new log levels to dump FIXMEs into the log and to log data
in form of a hex dump (#578114).
API: GST_CAT_FIXME_OBJECT
API: GST_CAT_MEMDUMP_OBJECT
API: GST_CAT_FIXME
API: GST_CAT_MEMDUMP
API: GST_FIXME_OBJECT
API: GST_MEMDUMP_OBJECT
API: GST_FIXME
API: GST_MEMDUMP
Stefan Kost [Wed, 8 Apr 2009 15:13:42 +0000 (18:13 +0300)]
docs: xref more
Stefan Kost [Wed, 8 Apr 2009 14:49:18 +0000 (17:49 +0300)]
tests: remove the hacks to workaround the pad-leak
Stefan Kost [Wed, 8 Apr 2009 12:24:58 +0000 (15:24 +0300)]
padtemplate: enable code to fix the leak, now that the deps have been released
Good and ffmpeg are actually multiple releases beyond, so that this is now safe
to do.
Felipe Contreras [Sat, 4 Apr 2009 18:18:23 +0000 (21:18 +0300)]
Automatic update of common submodule
From d0ea89e to b3941ea
Edward Hervey [Sat, 4 Apr 2009 12:53:21 +0000 (14:53 +0200)]
Automatic update of common submodule
From f8b3d91 to d0ea89e
Edward Hervey [Sat, 4 Apr 2009 12:42:04 +0000 (14:42 +0200)]
gst-inspect: remove dead assignment
first_flag will be either:
* rewritten without being read if we loop again (line 284)
* not read again if we don't loop
Edward Hervey [Sat, 4 Apr 2009 12:39:51 +0000 (14:39 +0200)]
basesink: Remove dead assignments.
sstart/sstop/rstart/rstop are all either:
* assigned values later on before being used in 'do_times:' (EOS and buffers)
* not used (non-EOS events)
Edward Hervey [Sat, 4 Apr 2009 12:38:52 +0000 (14:38 +0200)]
basesrc: remove dead assignment.
The variable will not be read before it's assigned a value line 942/945
Edward Hervey [Sat, 4 Apr 2009 12:37:13 +0000 (14:37 +0200)]
gsttaglist: Remove unused variable.
We don't need to allocate a variable if it's the return of a function call
and we only check it once.
Edward Hervey [Sat, 4 Apr 2009 12:35:34 +0000 (14:35 +0200)]
gststructure: Only use methods used in g_* checks if glib checks are disabled
Sebastian Dröge [Sat, 4 Apr 2009 08:59:39 +0000 (10:59 +0200)]
gst: Use g_once_init* or G_DEFINE_TYPE
Sebastian Dröge [Sat, 4 Apr 2009 08:20:36 +0000 (10:20 +0200)]
gst: Use G_DEFINE_TYPE and friends or at least g_once_init_* in the _get_type() functions
Sebastian Dröge [Sat, 4 Apr 2009 08:18:42 +0000 (10:18 +0200)]
gst: Use G_DEFINE_TYPE and don't call g_thread_init() from class_init
class_init is too late for calling g_thread_init() as g_thread_init()
needs to be called before any GObject function.
Mark Nauwelaerts [Fri, 3 Apr 2009 11:46:18 +0000 (13:46 +0200)]
Use g_slice_copy instead of g_slice_dup.
A (buggy) glib g_slice_dup macro may cause compiler warnings on e.g. x86_64.
Edward Hervey [Fri, 3 Apr 2009 10:21:55 +0000 (12:21 +0200)]
controller: remove dead assignment.
The value of prop is being overwritten just after without being read.
Edward Hervey [Fri, 3 Apr 2009 10:20:36 +0000 (12:20 +0200)]
gststructure: Remove dead assignment.
'type' is never used until line 1847 where it's overwritten.
Edward Hervey [Fri, 3 Apr 2009 10:19:40 +0000 (12:19 +0200)]
adapter: remove dead assignment.
The value set to to_copy at that line is never used, and is overwritten
further down before being read.
Edward Hervey [Fri, 3 Apr 2009 10:17:33 +0000 (12:17 +0200)]
gstbin: Remove unused variable.
The return value of gst_element_change_state isn't used after that call.
Edward Hervey [Fri, 3 Apr 2009 10:15:38 +0000 (12:15 +0200)]
pipeline: remove redundant assignment.
If that block is entered, then start_time becomes GST_CLOCK_TIME_NONE.
Since start_time is invalid, the code will enter the block at line 434 and
new_base_time will be set there.
Edward Hervey [Fri, 3 Apr 2009 10:13:38 +0000 (12:13 +0200)]
gstregistrybinary: remove variable only used for a check.
that variable isn't used anywhere else within that block.
Edward Hervey [Fri, 3 Apr 2009 10:13:00 +0000 (12:13 +0200)]
basesink : Remove unused variable.
sync is never used anywhere in that code.
Edward Hervey [Fri, 3 Apr 2009 10:12:08 +0000 (12:12 +0200)]
basetransform: move unused variable in the #if 0 block.
That variable is only used by the code which has been if 0'd
Edward Hervey [Fri, 3 Apr 2009 09:56:48 +0000 (11:56 +0200)]
Remove unused increments as detect by LLVM's CLang static analyzer.
Edward Hervey [Fri, 3 Apr 2009 09:52:49 +0000 (11:52 +0200)]
Remove unused variables detected by LLVM's Clang static analyzer.
Tim-Philipp Müller [Fri, 3 Apr 2009 10:19:42 +0000 (11:19 +0100)]
docs: improve API reference for gst_caps_get_structure()
Thomas Vander Stichele [Thu, 2 Apr 2009 11:32:58 +0000 (13:32 +0200)]
docs: explain ref ownership for handle_message implementations
Sebastian Dröge [Thu, 2 Apr 2009 08:43:16 +0000 (10:43 +0200)]
gstcheck: Call gst_check_init() before creating the suite
This allows using the GStreamer or GObject API in the suite
creation function.
Thiago Santos [Tue, 31 Mar 2009 21:14:08 +0000 (18:14 -0300)]
gst-launch: Fixes error when DISABLE_FAULT_HANDLER is defined
When defined, this macro prevented the declaration of 'waiting_eos', causing an error.
Edward Hervey [Thu, 26 Mar 2009 16:25:08 +0000 (17:25 +0100)]
capsfilter. Always calls _suggest, even with NULL caps. Fixes #574805
Sebastian Dröge [Mon, 30 Mar 2009 13:45:02 +0000 (15:45 +0200)]
tests: Don't define global buffers variable, it's already defined by libgstcheck
Peter Kjellerstedt [Mon, 30 Mar 2009 08:33:51 +0000 (10:33 +0200)]
docs: Some grammar and typo corrections.
Thomas Vander Stichele [Sun, 29 Mar 2009 11:41:22 +0000 (13:41 +0200)]
Fix typo.
Stefan Kost [Fri, 27 Mar 2009 15:30:23 +0000 (17:30 +0200)]
binaryregistry: init variable, that is referenced in error case below the fail: label
Wim Taymans [Fri, 27 Mar 2009 15:15:55 +0000 (16:15 +0100)]
clock: wakeup the async thread a bit more
Also wake up the async thread when it is doing an async wait for an entry.
Wim Taymans [Fri, 27 Mar 2009 15:15:10 +0000 (16:15 +0100)]
element: Fix a little debug message
Stefan Kost [Thu, 26 Mar 2009 11:07:56 +0000 (13:07 +0200)]
binaryregistry: check for not reading beyond the data area. Fixes #576842
Check all reads against the end of the data region. Roll back registration of
partial reads.
Stefan Kost [Wed, 25 Mar 2009 09:03:22 +0000 (11:03 +0200)]
docs: add a page about building gstreamer and apps
Thiago Santos [Thu, 26 Mar 2009 16:08:01 +0000 (13:08 -0300)]
Adds flag for eos on shutdown in gst-launch. Fixes #575814.
Wim Taymans [Thu, 26 Mar 2009 21:05:31 +0000 (22:05 +0100)]
clock: make UNSCHEDULED checks threadsafe
Move the checks for using an unscheduled entry from the unsafe GstClock to the
SystemClock object so that we can perform the correct locking.
fix a leak and potential deadlock then the async thread fails to start.
Sprinkle some G_LIKELY around because we can.
Wim Taymans [Thu, 26 Mar 2009 20:40:20 +0000 (21:40 +0100)]
clock: remove pending async wakeup sooner
Remove a pending async wakeup before we check if the next entry is UNSCHEDULED
because we might leave the control socket busy.
Peter Kjellerstedt [Thu, 26 Mar 2009 18:33:41 +0000 (19:33 +0100)]
gstpoll: Corrected a documentation typo.
Wim Taymans [Thu, 26 Mar 2009 18:13:55 +0000 (19:13 +0100)]
clock: add some more comments.
Wim Taymans [Thu, 26 Mar 2009 17:46:35 +0000 (18:46 +0100)]
clock: rework the wakeup of entries.
Keep a counter for the amount of outstanding wakeups that we produce and only
perform a write/read to the control socket when 1 or 0 respectively.
don't poll when waiting for the entries to be unblocked and clear their wakeup
counts, just act on the signal when the wakeup count is 0.
unscheduled entries will clear their wakeup count themselves.
Keep track of when we wakeup the async thread because the list of entries has
changed.
don't try to see if the list changed because we can't really know when one entry
is added multiple times.
Only wake up the async thread when we add an async entry to the head of the list
and the old entry was BUSY.
Jan Schmidt [Wed, 25 Mar 2009 17:31:16 +0000 (17:31 +0000)]
gstpoll: Fix up documentation strings.
Note the changed behaviour of gst_poll_wait for timer GstPoll's, and
fix a couple of spelling errors.
Wim Taymans [Thu, 26 Mar 2009 14:55:02 +0000 (15:55 +0100)]
clock: fix 2 wakeup races.
when an entry being waited on in the async thread is unscheduled, clear the
wakeup queue so we can continue waiting on other entries.
When an entry being waited on in the async thread is unlocked because an earlier
entry was added to the list, set the entry to OK again. This makes sure that
only the entries being waited on have the BUSY flag set and wake up the timer
poll when they are unscheduled.
Tim-Philipp Müller [Thu, 26 Mar 2009 14:44:04 +0000 (14:44 +0000)]
registry: ignore .git directory when recursively scanning plugin paths for plugins
Saves some cycles/pandas for those of us who run uninstalled setups.
Tim-Philipp Müller [Thu, 26 Mar 2009 14:16:55 +0000 (14:16 +0000)]
registry: do fsync() before close() and rename()
This helps prevent filesystem/data inconsistencies in certain
circumstances on certain filesystems (like ext4, xfs, ubifs).
Also see bug #562976.
Tim-Philipp Müller [Thu, 26 Mar 2009 01:09:03 +0000 (01:09 +0000)]
API: add gst_tag_list_get_buffer{_index}
Convenience API, mostly for image tags, so people don't have to
figure out the whole GValue/GstValue thing just for this.
Jan Schmidt [Wed, 25 Mar 2009 23:03:38 +0000 (23:03 +0000)]
systemclock: Clean up the tests a bit.
Add some cleanups to the system clock tests, to free all the memory and
unschedule/unref all clock IDs we allocate.
Use a mutex in one test to avoid potential threading problems on multicore
machines.
Jan Schmidt [Wed, 25 Mar 2009 21:37:38 +0000 (21:37 +0000)]
systemclock: Add a test for sync/async clockid interactions
This test randomly hangs if there are problems with the reliability of
unscheduling sync and async clockID's on the system clock.
Sebastian Dröge [Thu, 26 Mar 2009 10:17:01 +0000 (11:17 +0100)]
segment: Use g_slice_dup() now
Sebastian Dröge [Thu, 26 Mar 2009 10:08:27 +0000 (11:08 +0100)]
Remove some compatibility stuff for GLib < 2.14
Tim-Philipp Müller [Wed, 25 Mar 2009 00:50:07 +0000 (00:50 +0000)]
API: add GST_TAG_SUBTITLE_CODEC
Yes, 'codec' isn't exactly the best word, but let's be consistent with AUDIO_CODEC
and VIDEO_CODEC (which may be 'raw' formats as well after all). Prerequisite for
bug #576552.
Tim-Philipp Müller [Tue, 24 Mar 2009 21:39:21 +0000 (21:39 +0000)]
docs: gst-launch man page fix
The command line option is --gst-debug-disable, not --gst-disable-debug.
Fixes #576556. Spotted by Bogdan Harjoc.
Wim Taymans [Tue, 24 Mar 2009 18:33:56 +0000 (19:33 +0100)]
gstutils: improve property set and convert code
Use string deserialisation instead of custom parsing code to allow for all
supported ways of specifying property values.
fixes #576582.
Stefan Kost [Mon, 23 Mar 2009 13:18:21 +0000 (15:18 +0200)]
build: define stubs when disabling gst-debug subsystem. Fixes #575922
Running configure with e.g. --disable-dst-debug was compiling out the debug
system (ABI break). Now stubs are added and only if one does e.g.
make CFLAGS="-DGST_REMOVE_DISABLED" the symbols are ommitted.
Tim-Philipp Müller [Mon, 23 Mar 2009 12:34:34 +0000 (12:34 +0000)]
basesink: fix once-per-second 'emergency rendering' for case where all buffers but the very first are late
Due to a typo basesink didn't do any emergency rendering of late buffers
if the only buffer ever rendered was the first one with timestamp 0. This
means that in cases where the decoder is very very slow, we'd never see
any buffers but the very first one rendered. Fixes #576381.
Jan Schmidt [Sat, 21 Mar 2009 02:34:04 +0000 (02:34 +0000)]
docs: tweak the release procedure script
LRN [Fri, 20 Mar 2009 13:12:55 +0000 (14:12 +0100)]
win32: fix seeking in files >4GB
Use 64-bit functions on windows to implement seeking in files bigger
than 4GB.
Fixes #575988
Stefan Kost [Fri, 20 Mar 2009 09:26:30 +0000 (11:26 +0200)]
controller: Fix generation of control-change arrays.
When generating arrays of control changes timestamp variable was used instead
the local ts variable that we increment when stepping through the array.
Pointed out by Martin Pokorny.